Solar street lamp operation state monitoring method and solar street lamp

Abstract

The invention relates to the field of street lamp operation monitoring, in particular to a solar street lamp operation state monitoring method and a solar street lamp, and the method comprises the steps: carrying out the clustering of street lamps according to the environment parameters of the street lamps, and obtaining a plurality of clusters; calculating the generating capacity outlier degree of the target street lamp; when the outlier degree is greater than a preset first threshold value, generating a monitoring signal I, otherwise, generating a monitoring signal II; calculating the change degree of the generating capacity similarity between each clustering cluster and the adjacent clustering cluster; when the change degree is greater than a preset second threshold value, generating a monitoring signal III, otherwise, generating a monitoring signal IV; responding to the monitoring signal I and the monitoring signal IV, and judging that the operation state of the target street lamp individual is abnormal; and in response to the monitoring signal II and the monitoring signal III, judging that the street lamps in the corresponding region of the clustering cluster have regional abnormality. The method has the effect of improving the abnormal judgment accuracy of the running state of the street lamp.

Description

Technical Field

[0001] This application relates to the field of street lamp operation monitoring, and particularly to a method and system for monitoring the operation status of solar street lamps. Background Art

[0002] The core components of a solar street lamp include a solar panel, a storage battery, and a lamp body. During the day, the solar panel converts light energy into electrical energy and stores it in the storage battery. At night, the storage battery powers the street lamp. Solar street lamps are usually installed outdoors, and the power generation of the solar panel is related to the amount and intensity of the light received by the solar panel. If the solar panel is covered, for example, blocked by dust, sand, or snow from direct sunlight hitting the photovoltaic cells, this will cause a decrease in the photoelectric conversion efficiency of the solar panel, reduce power generation, and the storage battery of the street lamp cannot be fully charged, which may lead to a shortened lighting time at night or even the inability to light up.

[0003] By monitoring the power generation of the solar panel, it is possible to timely detect whether there is an abnormality in power generation. Through the Internet of Things technology, the operation data of the solar street lamp is uploaded to the cloud to achieve remote monitoring and management. When an abnormal situation (such as a decrease in power generation) is detected, the system automatically sends an alarm and provides maintenance suggestions.

[0004] However, the power generation of solar street lamps is affected by various factors, including seasonal changes, weather conditions, geographical location, and the lighting time period of the day. For example, in winter, due to the short sunshine time and low solar altitude angle, the power generation of the solar panel is usually lower than that in summer; and on cloudy, rainy, or hazy days, the power generation will also decrease significantly. Therefore, using a fixed threshold may not accurately reflect the actual operating conditions. If a fixed power generation threshold is set, it may lead to misjudgment or missed judgment of abnormal situations. Summary of the Invention

[0005] To solve the technical problem of low accuracy in determining abnormal power generation of solar street lamps, this application provides a method and system for monitoring the operation status of solar street lamps.

[0006] In a first aspect, this application provides a method for monitoring the operation status of solar street lamps, adopting the following technical solution:

[0007] Solar street lamp operation status monitoring method, including steps: clustering street lamps according to environmental parameters of the street lamps to obtain multiple clustering clusters; taking any street lamp in the clustering cluster as a target street lamp, and calculating the degree of outlier of the power generation of the target street lamp; in response to the degree of outlier being greater than a preset first threshold, generating a monitoring signal one, otherwise generating a monitoring signal two; calculating the degree of change in the power generation similarity between each clustering cluster and adjacent clustering clusters; in response to the degree of change being greater than a preset second threshold, generating a monitoring signal three, otherwise generating a monitoring signal four; in response to monitoring signal one and monitoring signal four, determining that the individual operation status of the target street lamp is abnormal; in response to monitoring signal two and monitoring signal three, determining that there is a regional abnormality in the street lamps in the area corresponding to the clustering cluster; in response to monitoring signal one and monitoring signal three, determining that there is a regional abnormality in the street lamps in the area corresponding to the clustering cluster and the individual operation status of the target street lamp is abnormal; otherwise, determining that the street lamp status is normal.

[0008] Optionally, snow parameters and particulate matter parameters, the particulate matter parameters include particulate matter concentration, wind speed, humidity and rainfall, and the snow parameters include temperature and snowfall.

[0009] Optionally, the degree of outlier of power generation is the deviation of the real-time power generation of the target street lamp from the real-time power generation of the street lamps within the cluster where it is located; or the degree of outlier of power generation is the deviation degree of the real-time power generation of the target street lamp from its historical power generation.

[0010] Optionally, the expression of the degree of outlier of power generation is: In the formula, L represents the degree of outlier of the power generation of the target street lamp, n represents the total number of street lamps in the cluster where the target street lamp is located, g i represents the actual power generation of the target street lamp at the i-th moment, represents the average power generation of other street lamps in the cluster where the target street lamp is located at the i-th moment, and σ represents the standard deviation of the power generation of other street lamps in the cluster where the target street lamp is located.

[0011] Optionally, the expression of the degree of outlier of power generation is: In the formula, L represents the degree of outlier of the power generation of the target street lamp, n represents the total number of street lamps in the cluster where the target street lamp is located, g i represents the actual power generation of the target street lamp at the i-th moment, G represents the mean value of the historical power generation of the target street lamp, σ ′ represents the standard deviation of the historical power generation of the target street lamp, and norm represents the standard normalization function.

[0012] Optionally, the calculation method of power generation similarity is: for the average power generation sequence of each clustering cluster; taking the normalized result of the cosine similarity of the corresponding power generation sequences of two clustering clusters as the power generation similarity.

[0013] Optionally, the calculation formula of power generation similarity is: Among them, S represents the similarity of power generation, is the i-th value in the power generation sequence of the clustering center of cluster one, is the i-th value in the power generation sequence of the clustering center of cluster two.

[0014] Optionally, the expression for the degree of change in the similarity of power generation is: Among them, D represents the degree of change in the similarity of power generation, S represents the sequence of similarities of power generation at historical moments of two clusters, max represents the maximum value function, min represents the minimum value function, and mean represents the average value function.

[0015] Optionally, the clustering algorithm is DBSCAN clustering or K-Means clustering.

[0016] In a second aspect, the present application provides a solar street lamp, adopting the following technical solution:

[0017] The solar street lamp monitors the power generation through the above-mentioned method for monitoring the operating state of the solar street lamp.

[0018] The present application has the following technical effects:

[0019] By clustering street lamps according to environmental parameters (such as snow accumulation parameters, particulate matter parameters, etc.), street lamps with similar environmental conditions can be grouped into the same cluster. This clustering method can more accurately reflect the actual operating environment of the street lamps, thereby avoiding misjudgment caused by environmental differences. For example, in a dusty or snow-covered area, even if the power generation is low, it may be a normal phenomenon and will not be wrongly marked as abnormal.

[0020] Supports dual monitoring of individual and regional anomalies. Compare the degree of deviation of the power generation of a street lamp from other street lamps within the cluster. If the degree of deviation is high, the street lamp may be abnormal; at this time, it may be a fault of the street lamp; but if the street lamps within the cluster are all covered with dust or snow, there will be a situation where the degree of deviation of the power generation of a single street lamp is low. At this time, all the street lamps within the cluster have power generation faults.

[0021] Traditional monitoring methods may only rely on the power generation data of a single street lamp and are easily affected by environmental factors, resulting in false alarms. However, the present application can more comprehensively evaluate the state of the street lamp by combining intra-cluster comparison and inter-cluster comparison, thereby effectively reducing the false alarm rate. Description of the Drawings

[0022] Figure 1 is the flowchart of the method for monitoring the operating state of the solar street lamp in the embodiment of the present application. Detailed Embodiment

[0023] The embodiments of the present application disclose a method for monitoring the operating state of solar street lamps. Refer to Figure 1 , which includes steps S1 - S4, specifically as follows:

[0024] S1: Cluster the street lamps according to the environmental parameters of the street lamps to obtain multiple clusters.

[0025] Clustering is a process of grouping street lamps according to their environmental parameters. By classifying street lamps with similar environmental conditions into the same cluster, the operating state of the street lamps can be evaluated more accurately.

[0026] Environmental parameters are important factors affecting the power generation of solar street lamps. To achieve accurate clustering, the present application selects environmental parameters closely related to the operating state of the street lamps. In one embodiment, the environmental parameters include: snow accumulation parameters and particulate matter parameters. The particulate matter parameters include particulate matter concentration, wind speed, humidity, and rainfall. The environmental parameters can also reflect the location information of the street lamps. When the environmental parameters in a region are relatively similar, it indicates that the street lamps in this region are under similar environmental conditions, and thus it can be inferred that these street lamps have similar location information or geographical distribution characteristics.

[0027] The clustering algorithm is DBSCAN clustering or K-Means clustering. The clustering algorithm can be the K-means clustering algorithm or the DBSCAN clustering algorithm. The K-means algorithm requires specifying the number of clusters in advance. However, in practical applications, the number of clusters cannot be accurately set, and the accuracy of the clustering result is relatively low. Therefore, the present application preferably uses the DBSCAN clustering algorithm, which can automatically identify the number of clusters. The process of clustering is a prior art and will not be elaborated here. After clustering, multiple clusters are obtained.

[0028] S2: Take any street lamp in the cluster as the target street lamp, and calculate the degree of deviation of the power generation of the target street lamp; in response to the degree of deviation being greater than a preset first threshold, generate monitoring signal one, otherwise generate monitoring signal two.

[0029] After clustering, each cluster contains several street lamps. To evaluate whether the operating state of a certain street lamp (target street lamp) is abnormal, it is necessary to calculate the deviation degree of its power generation from the power generation of other street lamps in the cluster, that is, the degree of deviation of power generation. If the degree of deviation of the power generation of the target street lamp exceeds the preset first threshold, it is considered that the street lamp may have a fault or abnormality, and monitoring signal one is generated. If the degree of deviation of the power generation of the target street lamp is lower than or equal to the first threshold, it is considered that the street lamp is operating normally, and monitoring signal two is generated.

[0030] In one embodiment, the degree of deviation of power generation is the deviation of the real-time power generation of the target street lamp from the real-time power generation of the street lamps within the cluster to which it belongs; since the street lamps within the same cluster have similar environmental conditions, it can be considered that their power generations should be close. If the power generation of the target street lamp significantly deviates from the average value of the other street lamps within the cluster, it may indicate that there is an abnormality with the street lamp.

[0031] Specifically, the expression for the degree of deviation of power generation is: In the formula, L represents the degree of deviation of power generation of the target street lamp, n represents the total number of street lamps in the cluster to which the target street lamp belongs, g i represents the actual power generation of the target street lamp at the i-th moment, represents the average power generation of the other street lamps in the cluster to which the target street lamp belongs at the i-th moment, and σ represents the standard deviation of the power generations of the other street lamps in the cluster to which the target street lamp belongs. If the degree of deviation of power generation is large, it indicates that the power generation of the target street lamp significantly deviates from the average value of the other street lamps within the cluster and there may be an abnormality. On the contrary, it indicates that the power generation of the target street lamp is close to that of the other street lamps within the cluster and the operating state is normal.

[0032] In addition to the method based on comparison within the cluster, the degree of deviation of the target street lamp can also be evaluated by comparing its real-time power generation with its historical power generation. In one embodiment, the degree of deviation of power generation can also be the degree of deviation of the real-time power generation of the target street lamp from its historical power generation.

[0033] Specifically, the expression for the degree of deviation of power generation is: In the formula, L represents the degree of deviation of power generation of the target street lamp, n represents the total number of street lamps in the cluster to which the target street lamp belongs, g i represents the actual power generation of the target street lamp at the i-th moment, G represents the mean value of the historical power generation of the target street lamp, and σ ′ represents the standard deviation of the historical power generation of the target street lamp, and norm represents the standard normalization function. If L is large, it indicates that the real-time power generation of the target street lamp significantly deviates from its historical power generation level and there may be an abnormality. On the contrary, it indicates that the real-time power generation of the target street lamp is close to its historical power generation level and the operating state is normal.

[0034] Comparison within the cluster is suitable for evaluating the relative state of the target street lamp and the surrounding street lamps and can effectively exclude the influence of environmental factors. Comparing with historical data is suitable for evaluating the long-term change trend of the target street lamp itself and can detect equipment aging or other systematic problems.

[0035] Exemplarily, the first threshold is set to 0.8. The first threshold is used to determine whether the power generation of the target street lamp significantly deviates from the other street lamps within its cluster or its own historical power generation level.

[0036] In practical applications, it is usually assumed that the power generation of street lights within a cluster has a small difference (low standard deviation). Therefore, when the degree of deviation of the power generation of a certain street light exceeds 0.8, it can be considered that its deviation degree is significant. This means that the deviation of the power generation of the target street light from the average power generation of other street lights within the cluster exceeds 0.8 times the standard deviation within the cluster. Such a deviation may indicate that there are problems such as occlusion, damage, or other issues with the street light.

[0037] Historical power generation reflects the power generation level of street lights under normal operating conditions. If the deviation between the current power generation and the historical mean exceeds 0.8 times the standard deviation, it may indicate a decline in the performance of the street light or the influence of external factors (such as snow cover or dust accumulation).

[0038] S3: Calculate the degree of change in the power generation similarity between each cluster and adjacent clusters; in response to the degree of change being greater than a preset second threshold, generate monitoring signal three, otherwise generate monitoring signal four.

[0039] After clustering is completed, each cluster contains several street lights with similar environmental conditions. To evaluate whether there are significant changes in power generation between different clusters, it is necessary to calculate the degree of change in power generation similarity. If the degree of change exceeds the preset second threshold, it is considered that there is a significant difference in power generation between the two clusters, and monitoring signal three is generated. If the degree of change is lower than or equal to the second threshold, it is considered that the change in power generation between the two clusters is within the normal range, and monitoring signal four is generated.

[0040] In one embodiment, the calculation method of power generation similarity is as follows: for the average power generation sequences of each cluster; the normalized result of the cosine similarity of the corresponding power generation sequences of the two clusters is used as the power generation similarity. Cosine similarity is a commonly used similarity measurement method for measuring the cosine value of the angle between two vectors. The smaller the angle, the higher the similarity. Here, the average power generation sequences of the two clusters are regarded as two vectors, and the cosine similarity between them is calculated to evaluate the similarity of power generation. Since the result of cosine similarity is usually in the range of [-1, 1], it is necessary to normalize it (such as mapping it to the range of [0, 1]) for unified comparison with other metrics.

[0041] In one embodiment, the calculation formula of power generation similarity is: Among them, S represents the power generation similarity, is the i-th value in the power generation sequence of the clustering center of cluster one, is the i-th value in the power generation sequence of the clustering center of cluster two. If S is large (close to 1), it indicates that the power generation sequences of the two clusters are highly similar. If S is small (close to 0), it indicates that the power generation sequences of the two clusters are quite different.

[0042] The expression for the degree of change in power generation similarity is as follows: Among them, D represents the degree of change in power generation similarity, S represents the sequence of power generation similarity at historical moments of two clustering clusters, max represents the maximum value function, min represents the minimum value function, and mean represents the average value function. If D is relatively large, it indicates that there has been a significant change in the power generation similarity between the two clustering clusters, and there may be regional anomalies. On the contrary, the change in power generation similarity between the two clustering clusters is small, and the operating state is relatively stable.

[0043] The method for dividing adjacent clustering clusters is to define and delimit adjacent clustering clusters based on the distance between cluster centers being less than a distance threshold. The cluster center represents the average characteristics of all street lights in the cluster in the environmental parameter space. If the DBSCAN clustering algorithm is used, the cluster center can be the mean point of the environmental parameters of all street lights in the cluster. If the K-Means clustering algorithm is used, the cluster center is the centroid generated during the algorithm iteration process. In a relatively small area (such as a district or a city), a relatively small distance threshold (such as 0.1 or 0.2) can be selected. In a relatively large area (such as multiple cities or provinces), a relatively large distance threshold (such as 0.5 or 1.0) can be selected.

[0044] The second threshold is used to determine whether the change in power generation similarity between different clustering clusters is significant. Exemplarily, the second threshold can be set to 0.2. If the second threshold is greater than 0.2, it indicates that there has been a significant change in the power generation similarity between the two clusters, which may be due to changes in environmental conditions (such as snowfall, etc.) resulting in a general decrease in power generation in the entire region. On the contrary, it means that the change range of power generation similarity exceeds 20% of its average value. This change range can usually reflect the occurrence of regional anomalies.

[0045] In practical applications, the specific values of the first threshold and the second threshold can be adjusted according to the following factors:

[0046] In areas with high dust or particulate matter concentration, the power generation fluctuates greatly. Therefore, the first threshold can be appropriately increased to reduce false alarms. In cold or snowy areas, the power generation is greatly affected by snow cover, and the first threshold can also be appropriately increased. If the power generation distribution of street lights within the cluster is relatively uniform (with a small standard deviation), the first threshold can be set relatively low. If the power generation distribution of street lights within the cluster is relatively dispersed (with a large standard deviation), the first threshold needs to be set relatively high.

[0047] In a relatively small area (such as a district or a city), due to relatively consistent environmental conditions, the second threshold can be set relatively low. In a relatively large area (such as multiple cities or provinces), due to possible large differences in environmental conditions, the second threshold can be set relatively high.

[0048] S4: In response to Monitoring Signal 1 and Monitoring Signal 4, determine that the operating state of the target street lamp individual is abnormal; in response to Monitoring Signal 2 and Monitoring Signal 3, determine that there are regional abnormalities in the street lamps in the area corresponding to the clustering cluster; in response to Monitoring Signal 1 and Monitoring Signal 3, determine that there are regional abnormalities in the street lamps in the area corresponding to the clustering cluster and the operating state of the target street lamp individual is abnormal; conversely, determine that the street lamp state is normal.

[0049] When Monitoring Signal 1 and Monitoring Signal 4 are triggered simultaneously, the system will determine that a specific street lamp (the target street lamp individual) has an operating abnormality. The simultaneous occurrence of these two signals may indicate problems with the power supply system or the light source itself of this street lamp.

[0050] When Monitoring Signal 2 and Monitoring Signal 3 are triggered simultaneously, the system will determine that there are regional abnormalities in multiple street lamps within a certain area (the clustering cluster). The simultaneous occurrence of these two signals may indicate that the solar panels in this area are blocked (such as continuous rainy days or tree shading), the storage batteries are aging, or other regional problems.

[0051] When Monitoring Signal 1 and Monitoring Signal 3 are triggered simultaneously, the system will not only determine that a specific street lamp (the target street lamp individual) has an operating abnormality, but also determine that there are overall regional abnormalities in the street lamps within the area (the clustering cluster) to which this street lamp belongs. This situation may indicate that problems with the storage batteries in this area have caused some street lamps (including the target street lamp) to fail to work properly, and at the same time, the target street lamp also has additional individual faults.

[0052] If none of the above situations occur, the system determines that the status of all street lamps is normal. The solar panels generate electricity normally, the storage batteries have sufficient power, the street lamp bulbs work normally, and no abnormal signals are triggered.

[0053] The embodiments of this application also disclose a solar street lamp for monitoring the power generation amount by the method for monitoring the operating state of the solar street lamp. The solar street lamp is equipped with an environmental parameter acquisition module for real-time acquisition of environmental data related to the operating state of the street lamp. These environmental parameters include, but are not limited to, snow parameters (such as temperature, snowfall amount) and particulate matter parameters (such as particulate matter concentration, wind speed, humidity, rainfall). By collecting these parameters, the actual operating environment where the street lamp is located can be more accurately evaluated, thereby providing basic data support for subsequent clustering analysis and anomaly detection. The solar street lamp is built-in with a power generation amount monitoring module capable of real-time recording of the power generation amount data of each street lamp. At the same time, through embedded algorithms or cloud computing platforms, outlier degree analysis and similarity change analysis are performed on the power generation amount data.

[0054] Solar street lights use built-in intelligent algorithms (such as DBSCAN clustering or K-Means clustering) to divide street lights into multiple clustering clusters according to environmental parameters. Street lights within each cluster have similar environmental conditions and operating characteristics. Based on the clustering results, the street lights can automatically determine whether their operating status is normal and generate corresponding monitoring signals.

[0055] Solar street lights can upload monitoring data to the cloud platform through a wireless communication module (such as LoRa, NB-IoT, etc.) to achieve remote monitoring and centralized management. Maintenance personnel can view the operating status of each street light in real time through the cloud platform and quickly locate problem street lights or problem areas based on the monitoring signals. In addition, the system also supports generating maintenance work orders to guide maintenance personnel in carrying out targeted repairs or cleaning work.

[0056] The above are all preferred embodiments of this application, and the protection scope of this application is not limited by this. Therefore, all equivalent changes made according to the structure, shape, and principle of this application should be covered within the protection scope of this application.

Claims

1. A method for monitoring the operating state of a solar street lamp, characterized in that, Including the steps: Clustering street lamps according to the environmental parameters of the street lamps to obtain multiple clustering clusters; Taking any street lamp in the clustering cluster as the target street lamp, calculating the degree of deviation of the power generation of the target street lamp; in response to the degree of deviation being greater than a preset first threshold, generating a first monitoring signal, otherwise generating a second monitoring signal; Calculating the degree of change in the power generation similarity between each clustering cluster and the adjacent clustering clusters; in response to the degree of change being greater than a preset second threshold, generating a third monitoring signal, otherwise generating a fourth monitoring signal; In response to the first monitoring signal and the fourth monitoring signal, determining that the individual operating state of the target street lamp is abnormal; In response to the second monitoring signal and the third monitoring signal, determining that there is a regional abnormality in the street lamps in the area corresponding to the clustering cluster; In response to the first monitoring signal and the third monitoring signal, determining that there is a regional abnormality in the street lamps in the area corresponding to the clustering cluster and the individual operating state of the target street lamp is abnormal; otherwise, determining that the street lamp state is normal.

2. The method for monitoring the operating state of a solar street lamp according to claim 1, wherein The environmental parameters include: snow accumulation parameters and particulate matter parameters. The particulate matter parameters include particulate matter concentration, wind speed, humidity, rainfall, and the snow accumulation parameters include temperature and snowfall.

3. The method for monitoring the operating state of a solar street lamp according to claim 1, wherein The degree of deviation of power generation is the deviation amount of the real-time power generation of the target street lamp from the real-time power generation of the street lamps within the cluster where it is located; or the degree of deviation of the power generation of the target street lamp from its historical power generation.

4. The method for monitoring the operating state of a solar street lamp according to claim 3, wherein The expression for the degree of outlier of power generation is as follows: In the formula, L represents the degree of outlier of the power generation of the target street lamp, n represents the total number of street lamps in the cluster where the target street lamp is located, and g i represents the actual power generation of the target street lamp at time i, represents the average power generation of other street lamps in the cluster where the target street lamp is located at time i, and σ represents the standard deviation of the power generation of other street lamps in the cluster where the target street lamp is located.

5. The method for monitoring the operating state of a solar street lamp according to claim 3, characterized in that The expression for the degree of outlier of power generation is as follows: In the formula, L represents the degree of outlier of the power generation of the target street lamp, n represents the total number of street lamps in the cluster where the target street lamp is located, and g i represents the actual power generation of the target street lamp at time i, G represents the mean value of the historical power generation of the target street lamp, and σ ′ represents the standard deviation of the historical power generation of the target street lamp, and norm represents the standard normalization function.

6. The method for monitoring the operating state of a solar street lamp according to claim 1, wherein The calculation method of power generation similarity is: for the average power generation sequence of each clustering cluster; taking the normalized result of the cosine similarity of the corresponding power generation sequences of two clustering clusters as the power generation similarity.

7. The method for monitoring the operating state of a solar street lamp according to claim 1, wherein The calculation formula for the power generation similarity is as follows: where S represents the power generation similarity, is the i-th value in the power generation sequence of the clustering center of cluster one, is the i-th value in the power generation sequence of the clustering center of cluster two.

8. The method for monitoring the operating state of a solar street lamp according to claim 1, wherein, The expression for the degree of change in power generation similarity is as follows: Among them, D represents the degree of change in power generation similarity, S represents the power generation similarity sequence at historical moments of two clustering clusters, max represents the maximum value function, min represents the minimum value function, and mean represents the average value function.

9. The method for monitoring the operating state of a solar street lamp according to any one of claims 1-8, characterized in that, The clustering algorithm is DBSCAN clustering or K-Means clustering.

10. Solar street lamp, characterized in that, Monitoring the power generation through the method for monitoring the operating state of a solar street lamp according to any one of claims 1-9.

Images